An AWS outage that began around 10:45am ET has led to disruptions for many services including Disney+, Tinder, Roku, Coinbase, Cash App, Amazon Music, and Venmo
Amazon says it is working toward a solution — Problems for some of Amazon's AWS cloud servers are causing slow loading …
Context & Ripple Effects
This is at least the third major AWS failure in recent years to trace back to the same infrastructure: an outage in November 2020 that took down Roku, Flickr, and Adobe, and a nearly four-hour US-EAST-1 outage in June 2023 that hit fast-food apps among others. The recurrence matters because each event re-exposes how much of the consumer internet — streaming, dating, payments, crypto — runs through a single Amazon region.
The blast radius also reaches inside Amazon itself: coverage of this same incident shows the outage knocked out the company's own Flex and Dolphin logistics apps, halting some deliveries. That makes this not just a customer-service story but a demonstration that even Amazon's operations are single-threaded on its own cloud.
First-order effects
- Users of Disney+, Tinder, Roku, Coinbase, Cash App, Amazon Music, and Venmo face slow loading or outright downtime, while those services' support teams absorb the complaints for an Amazon-side fault they cannot fix.
- Amazon's own delivery operations stall where Flex drivers and warehouse time-tracking depend on Dolphin, turning a cloud incident into physical logistics delays.
Second-order effects
- High-profile AWS customers like Coinbase and Venmo — where downtime touches money movement rather than just content — face renewed pressure from their own users and boards to justify or diversify their cloud concentration.
- Rival clouds gain a recurring sales argument: every US-EAST-1 failure gives Google Cloud and Microsoft's Azure teams a concrete case study when pitching multi-region or multi-cloud architectures to AWS's base.
Third-order effects
- If the pattern holds — the same region failing repeatedly across 2020, 2021, 2023, and again in the October 2025 US-EAST-1 outage — enterprise buyers will increasingly treat single-region AWS dependency as an architectural defect, pushing the industry toward multi-region defaults and multi-cloud hedging regardless of the added cost.
The trend: Recurring US-EAST-1 failures keep converting AWS's scale advantage into a visible systemic risk, steadily normalizing multi-region and multi-cloud design across the consumer internet.